Re: [Qemu-devel] [PATCH v4 11/11] qga-win: demystify namespace striping

[Eric Blake <eblake@redhat.com>]

On 10/4/18 8:20 AM, Marc-André Lureau wrote:
> Hi
> 
> On Thu, Oct 4, 2018 at 3:22 PM Tomáš Golembiovský <tgolembi@redhat.com> wrote:
>>
>> It was not obvious what exactly the cryptic string copying does to the
>> GUID. This change makes the intent clearer.

In the subject line, s/striping/stripping/ (this is about performing a 
'strip' operation on a prefix, but I read the subject as an instance of 
'stripe' as in drawing a line or fragmenting data in a RAID).

>> +++ b/qga/commands-win32.c
>> @@ -507,7 +507,14 @@ static GuestPCIAddress *get_pci_info(char *guid, Error **errp)
>>       char dev_name[MAX_PATH];
>>       char *buffer = NULL;
>>       GuestPCIAddress *pci = NULL;
>> -    char *name = g_strdup(&guid[4]);
>> +    char *name = NULL;
>> +
>> +    if ((g_str_has_prefix(guid, "\\\\.\\") == TRUE) ||
>> +        (g_str_has_prefix(guid, "\\\\?\\") == TRUE)) {

I find that 'cond == true' is redundant to just writing 'cond'. And that 
sentiment applies to both the <stdbool.h> 'bool' and to the glib 
abomination TRUE (why they had to invent their own boolean names, worse 
in every way compared to <stdbool.h>, is beyond me).

>> +        name = g_strdup(&guid[4]);
> 
> I find "guid + 4" easier to read though

Concur.

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3266
Virtualization:  qemu.org | libvirt.org